FT-IR for Rapid Discrimination of Halophilic Archaea and Bacteria

Fourier transform infrared spectroscopy (FT-IR) was applied for the first time for the discrimination of halophilic prokaryotes. In the current study, halophilic Bacteria and Archaea species were investigated according to their FT-IR profiles. Some of the halophilic isolates collected from Çamaltı Saltern in Turkey were also analyzed and their FT-IR profiles were compared with reference strains. Hierarchical cluster analysis was used to indicate discrimination. This study implies that FT-IR can be applied to discriminate species of Archae and Bacteria.


Introduction
Microorganisms requiring salt for growth and specialized for living in hypersaline environments are referred as halophiles. Kushner (1985) defined several categories of microorganisms according to the salt concentration that was optimal for growth. These organisms are phylogenetically (on the basis of comparative ribosomal RNA gene sequencing) found in all three domains of life: Archaea, Bacteria and Eukarya, the domain being the highest taxon. Two of these domains, the Archaea and Bacteria, are exclusively microbial and are composed of single cells that lack a membrane-enclosed nucleus (i.e., prokaryotic cells) [1].
Halophilic microorganisms able to live in saline environments have garnered a great deal of attention as they have the potential in various biotechnological applications such as the production of biopolymers, compatible solutes, cancer detection, drug screening, biodegradation of toxic compounds, fermentation of foods and most importantly the production of enzymes [2][3][4].
Most of the halophilic Archaea are found in hypersaline lakes and saltern crystallizer ponds, giving a red coloration due to C-50 carotenoid pigments. However, the domain Bacteria contains many types of halophilic and halotolerant microorganisms in many phylogenetic subgroups [5].
Although the domains Archaea and Bacteria are founded on phylogenetic criteria, these two domains can also be characterized by phenotypic properties, e.g. membrane lipids [6]. In contrast to the lipids of Bacteria in which ester linkages bond fatty acids to glycerol, the lipids of Archae contain ether bonds between glycerol and their hydrophobic side chains [7].
The diversity of Archaea and Bacteria in various hypersaline environments worldwide has been characterized by culture-based, pigment and molecular-phylogenetic studies [7][8][9][10][11][12]. It is known that halophilic Archaea are mostly orange-red pigmented and grow optimally at salt concentrations from above 20% [13]. Archaeal strains grow in the presence of penicillin G in the growth medium, whereas Bacterial strains could not [7]. However, the more recent technique using Archaeal and Bacterial primers in PCR has proved itself as a more reliable way of distinguishing between the two domains [14][15].
Conventional techniques for identifying microorganisms are phenotypic, biochemical or genotypic assays including culturing, immunological methods such as the enzyme-linked immunosorbent assay (ELISA), and molecular biology techniques such as polymerase chain reaction (PCR) and in-situ hybridization. These techniques are accurate, but none of these techniques are readily automated for rapid (<1 min) identification under industrial or field conditions [16]. However, the FT-IR method is rapid and non-invasive, permitting users to collect full spectra in a few seconds per sample. Differences in the structure and quantity of cell wall polysaccharides, lipids and proteins are reflected in the FT-IR spectra. These data provide fingerprint-like patterns which are highly reproducible and typical for different bacteria [17][18][19]. Compared to conventional methods, FT-IR identification is not only based on cell characteristics, but is achieved by selecting discriminating features from a large amount of spectral information. The technique is applicable to all microorganisms including prokaryotes and eukaryotes (e.g. fungi) that can be grown in culture [20].
Identification with FT-IR has been used for the identification of various members of Bacteria at the genus, species, and even strain level and typical examples are given in Table 1  . However, there has been no such study of Archaea except for one study by Hedrick et al. [6] who used diffuse reflectance FT-IR spectroscopic analysis for the rapid differentiation of Archaea from Bacteria. Therefore, in this study, we aimed to investigate the similarities and differences between some representatives of halophilic Archaea and Bacteria domains, and if possible to differentiate between genera and species using FT-IR spectroscopy. To our knowledge, this is the first study to investigate FT-IR differentiation of halophilic prokaryotes.

Organisms
A total of twelve halophilic reference strains representing Archaea and Bacteria domains were kindly provided by Josefa Anton (Spain), Michael Dyall-Smith (Australia), Helga Stan-Lotter (Austria) and Dr. Mehmet Burçin Mutlu (Anadolu University, Turkey). The strains used and their sources are given in Table 2.

Microorganisms
Year Reference   Nine of the Archeal and five of the Bacterial strains isolated from the Çamaltı solar saltern in Turkey for a previous study were also included in the assay (Table 2). Domain tests of these isolates were carried out by PCR amplification of 16S rRNA genes using universal primers for Bacteria and Archaea, as indicated by Mutlu et al [12]. Sample Preparation. Cells were collected and dried in a vacuum evaporator at room temperature and then were ground into fine particles using a mortar and pestle. About one milligram of biomass was encapsulated in 100 milligrams KBr with a pressure of 100 kg/cm 2 (2000 psi) in order to prepare the sample pellet. Three sample pellets were prepared for each microorganism tested.

Fourier transform infrared spectroscopy and data analysis
Infrared spectra were obtained by scanning the pellets with a Perkin Elmer Spectrum 100. The spectrum of air was recorded as the background and subtracted automatically using the spectrum software program. Atmospheric vapour was also automatically subtracted. The FT-IR spectra of bacterial samples were recorded in the 4000-450 cm -1 region at room temperature. Three spectra were acquired for each bacterial sample.
The averages of the spectra belonging to the same strain, baseline correction, normalization and the bands were obtained by the software. The average spectra and normalization process were applied only for visual representation of the differences; however, for the determination of the spectral parameters and calculation of mean values and statistical analyses, each baseline-corrected original spectrum was taken into consideration.

Statistical analysis
The features that reflect the important characteristics of different biochemical components was taken into account while assessing the significant values in spectra. Hierarchical cluster analysis was applied to classify the differences in samples by using Clustan 8.0 software. For calculating the spectral pair-wise distance, the Euclidian distance was used in the 4000-450 cm -1 region. Average Linkage (UPGMA) method was used to join the clusters together.

Results
In this study, FT-IR spectroscopy was carried out on halophilic members of the Archaea and Bacterial domains and spectral differences and similarities were compared between the domains and species.
The FT-IR spectra of reference strains of Archaea are shown in Figure 1. As can be seen from the figure, there were differences mainly in the 450-1700 cm -1 region. Table 3 shows the regions numbered from 1 to 6 to discriminate Archaeal members. All of the Archaeal strains showed ether bonds (C-O) in the 1000-1300 cm -1 region. Also, CH 2 asymmetric stretching of methylene groups in fatty acids was observed at 2930 cm -1 as broadened peaks (peak no. 1).
The FT-IR spectra of Bacteria strains are given in Figure 2. As can be seen in the figure, there were differences mainly in the 1000-1700 cm -1 region, and typically strong ester bonds in the 1735-1750 cm -1 region.  A strong peak indicating amide group N-H stretching at 3200 cm -1 (peak no. 1) was observed (Figure 2). CH 2 asymmetric stretching of methylene groups in fatty acids were observed at 2960 cm -1 as narrow peaks (peak no. 2-3). Table 4 shows regions 1 to 7 in Figure 2 used to discriminate Archaeal members. Hierarchical cluster analyses on FT-IR spectra from the reference Archaea and Bacteria strains resulted in the dendrogram shown in Figure 3.

Comparison of strains isolated from the camaltı saltern
The bands centered at 1000-1300 cm -1 corresponds to the C-O stretching of ether groups (Figure 4) for Archaeal strains obtained from the Çamaltı solar saltern. A strong band occurred at 1735-1750 cm -1 corresponding to C=O stretching of the esters ( Figure 5) in Bacterial strains obtained from the Çamaltı solar saltern. Hierarchical  The spectra were analyzed for the differentiation of each genus/ species. The FT-IR spectra for known examples of each class of bacteria were collected. These were used to generate a discrimination method based on the assignment of spectra to classes. As with any biological sample, there was considerable variation within each class, so region selection and spectral processing were critical.
Clear segregations between different domains, genera and species of bacteria were observed by hierarchical cluster analyses by using Clustan 8.0. Euclidean distances were taken into account.

Discussion
In this study, we used a spectroscopic approach for the rapid discrimination and identification of halophilic microorganisms. These extreme microorganisms are important for their newly discovered biotechnological purposes.     shows the dissimilarity level of strains tested by Clustan 8. Strains numbered from C15 to C-52 were isolated from Çamaltı saltern.
Traditional methods used for the identification of various microorganisms are based on morphological characteristics, biochemical reactions, serological reactions, sensitivity to bacteriophages and 16s rDNA sequencing. However, these methods can often be tedious, time consuming and occasionally inconclusive [38].
The only study on the rapid differentiation of archaebacteria from eubacteria by diffuse reflectance FT-IR spectroscopic analysis of lipid preparations was carried out by Hedrick et al. [6]; the authors concluded that archaebacterial ether lipids had broadened peaks in the alkyl region between 2800-3000 cm -1 ; however, eubacterial spectra were narrower and clearly separate [39]. It is known that ether bonds (C-O) appear in the 1000-1300 cm -1 region (http://www2ups.edu/ faculty/hanson/spectroscopy/IR/IRfrequencies.html).
We also observed that archaeal strains gave ether bands in this region; however, both reference and Çamaltı saltern bacterial strains had typically a strong ester bonds in the 1735-1750 cm -1 region.
The application of various vibrational spectroscopic techniques to identify and characterize microorganisms has been studied with Fourier transform infrared (FT-IR) spectroscopy are the most promising. FT-IR spectroscopy is a rapid and non-destructive technique that has been demonstrated to provide a highly sensitive and reproducible means for the identification of microorganisms without the use of reagents. Furthermore, it can provide results faster than virtually any other conventional method in microbiology, which are generally time consuming and labour intensive. Moreover, FT-IR identification methods have easy and simple protocols together with low running costs [50]. Infrared spectroscopy is based on the measurement of molecular vibrations [51]. Fourier transform infrared spectroscopy with high-resolution power is able to distinguish microbial cells at different taxonomic levels [52]. This technique is referred to as whole-organism fingerprinting because it measures the vibrations of chemical bonds within the biochemical constituents of cells [6]. The chemical composition of the cell contributes to the FT-IR spectrum [22]. As microorganisms consist of pure biochemicals, the IR spectra of these microorganisms provide few features and peaks since the bonds and functional groups of a particular molecular structure are often give a simple, unique, and reliable spectral fingerprint. The IR vibrations of microorganisms are related to thousands of other molecular components in microorganisms. With the use of different statistical methods like principal component analysis (PCA), hierarchical cluster analysis (HCA), partial least square regression (PLSR), soft independent modelling of class analogy (SIMCA), K-nearest neighbor (KNN), artificial neural networks (ANNs) and canonical variate analysis (CVA), extracting subtle details from the IR spectra of whole microbial samples has become possible. This method has been successfully applied to correlate the nearest neighbor along appropriate taxa [53][54]. In this study, clear segregations between different domain, genera and species of bacteria were observed by hierarchical cluster analyses.
In conclusion, identification of halophilic microorganisms using FT-IR seems to have great value, confirming that FT-IR can detect biochemical differences between halophilic prokaryotic cells in this preliminary study. However, future studies should be carried out to optimize the results by including many reference strains and environmental isolates.
This study has contributed basic data to the FT-IR studies of halophilic prokaryotes and the method could become a potentially useful tool for rapid differentiation of domains of Archaea and Bacteria at the strain level.